Isolation and characterization of human cDNA clones encoding the

Denis G. Bosc , Kevin C. Graham , Ronald B. Saulnier , Cunjie Zhang , David Prober .... Grazyna Dobrowolska , Flavio Meggio , Oriano Marin , Fred J. L...
1 downloads 0 Views 3MB Size
Download PDF
8436

Biochemistry 1990, 29, 8436-8447

Isolation and Characterization of Human cDNA Clones Encoding the a and the a’ Subunits of Casein Kinase IIty$ Fred J. L o z e m a n , ~ ~ l lDavid ~ L W. Litchfield,~*ll~i Carol Piening,g Koji Takio,§>#-*Kenneth A. Walsh,# and Edwin G. K r e b s * ~ ~ ~ l l ~ # Howard Hughes Medical Institute, Department of Pharmacology, and Department of Biochemistry, University of Washington, Seattle, Washington 98195 Received April 5, 1990; Revised Manuscript Received May 18, 1990

ABSTRACT: Casein kinase I1 is a widely distributed protein serine/threonine kinase. The holoenzyme appears to be a tetramer, containing two a or a’ subunits (or one of each) and two p subunits. Complementary DNA clones encoding the subunits of casein kinase I1 were isolated from a human T-cell XgtlO library using cDNA clones isolated from Drosophila melanoguster [Saxena et al. (1987) Mol. Cell. Biol. 7, 3409-34171. One of the human c D N A clones (hT4.1) was 2.2 kb long, including a coding region of 1 176 bp preceded by 156 bp (5’ untranslated region) and followed by 871 bp (3’ untranslated region). The hT4.1 clone was nearly identical in size and sequence with a c D N A clone from HepG2 human hepatoma cultured cells [Meisner et al. (1989) Biochemisrry 28, 4072-40761. Another of the human T-cell c D N A clones (hT9.1) was 1.8 kb long, containing a coding region of 1053 bp preceded by 171 bp (5’ untranslated region) and followed by 550 bp (3’ untranslated region). Amino acid sequences deduced from these two cDNA clones were about 85% identical. Most of the difference between the two encoded polypeptides was in the carboxy-terminal region, but heterogeneity was distributed throughout the molecules. Partial amino acid sequence was determined in a mixture of a and a’ subunits from bovine lung casein kinase 11. The bovine sequences aligned with the 2 human cDNA-encoded polypeptides with only 2 discrepancies out of 535 amino acid positions. This confirmed that the two human T-cell c D N A clones encoded the a and a’ subunits of casein kinase 11. Microsequence data determined from separated preparations of bovine casein kinase I1 a subunit and a’ subunit [Litchfield et al. (1990) J . Biol. Chem. 265, 7638-76441 confirmed that hT4.1 encoded the a subunit and hT9.1 encoded the a’ subunit. These studies show that there are two distinct catalytic subunits for casein kinase I1 ( a and a’) and that the sequence of these subunits is largely conserved between the bovine and the human.

c a s e i n kinase I1 is a protein kinase that catalyzes the phosphorylation of serine or threonine residues in proteins (Le., a protein serine/threonine kinase). The enzyme appears to be present in all eukaryotic cells [see reviews by Hathaway and Traugh (1982), Krebs et al. (1988), and Tuazon and Traugh (1990)], implying that it has fundamental cellular functions. Casein kinase I1 is referred to as a messenger-independent protein kinase because its activity is not affected by factors known to affect the activity of some other protein kinases in the cell (e.g., cyclic nucleotides, Ca2+,or diacylglycerols). The activity of casein kinase I1 is affected in vitro by a number of substances, including heparin (Hathaway & Traugh, 1982; Meggio et al., 1982), various polyamine compounds (Hara et al., 1981; Cochet & Chambaz, 1983a; Hathaway & Traugh, 1984a; Mamrack, 1989), 2,3-diphosphoglyceric acid (Hathaway & Traugh, 1984b; Gonzatti & Traugh, 1988), certain nucleic acids (Gatica et al., 1989), +This work was supported in part by Grant GM15731 from the National Institutes of Health (to K.A.W.). *The nucleic acid sequences in this paper have been submitted to GenBank under Accession Numbers 502920 and 502924. * To whom correspondence should be addressed at the Howard Hughes Medical Institute, Mail Stop SL-15, University of Washington, Seattle, WA 98195. Howard Hughes Medical Institute. Department of Pharmacology. Recipient of a fellowship from the Medical Research Council of Canada. # Department of Biochemistry. A Present address: Frontier Research Programs, RIKEN, Wako, Saitama, 351-01, Japan.

0006-2960 I90 10429-8436SO2.50I O

and a specific protein (Murao et al., 1989). The physiological significance of these effectors is unknown, but there has been speculation that at least some of them may help regulate the activity of casein kinase I1 in vivo (Hathaway & Traugh, 1984b; Gonzatti & Traugh, 1988). There is a large list of proteins that can be phosphorylated by casein kinase I1 in vitro, and a growing number of these appear to be substrates in vivo [see Hathaway and Traugh (1982), Sommercorn and Krebs (1987), Krebs et al. (1988), and Tuazon and Traugh (1 990)]. Some of the latter include acetyl-coA carboxylase (Witters et al., 1983, 1988; Haystead et a]., 1988), glycogen synthase (Picton et al., 1982; DePaoli-Roach et a]., 1983), c-myc protein (Liischer et al., 1988), the E7 protein of human papillovirus type 16 (Firzlaff et al., 1989), the large T-antigen protein of SV40 (Grasser et al., 1988), the 90-kDa heat shock proteins (Lees-Miller & Anderson, 1989), nucleolin (Caizergues-Ferrer et al., 1987; Schneider et al., 1988), and deoxyribonucleic acid (DNA)’ topoisomerase I1 (Ackerman et al., 1985, 1988). The putative cellular substrates listed above plus the many that are not listed include proteins involved in DNA replication, transcription, Abbreviations: ATP, adenosine 5’-triphosphate; bp, nucleotide bases; cDNA, complementary deoxyribonucleic acid: CNBr, cyanogen bromide; dATP, deoxyadenosine 5’-triphosphate; dITP, deoxyinosine 5’-triphosphate; DEAE, diethylaminoethyl: DME, Dulbecco-Vogt-modified Eagle’s; DNA, deoxyribonucleic acid; DNase, deoxyribonuclease; dpm, disintegrations per minute; FSBA, [p-(fluorosulfonyl)benzoyl]adenosine; h , hours; kb, nucleotide kilobases; LB, Luria-Bertani medium; min, minutes; mRNA, messenger ribonucleic acid; pfu, plaque forming unit(s); RNA, ribonucleic acid; RNase, ribonuclease; SDS, sodium dodecyl sulfate; TPCK, L- 1-(tosylamino)-2-phenylethyl chloromethyl ketone.

0 1990 American Chemical Societv

Sequence of Casein Kinase 11 a and

CY’ Subunits

translation, and the acute regulation of metabolism. This is especially interesting since the activity of casein kinase I1 is increased in extracts from cells treated with insulin or growth factors (Sommercorn et al., 1987; Klarlund & Czech, 1988; Carroll & Marshak, 1989; Ackerman & Osheroff, 1989), suggesting that casein kinase I1 is a regulatory component of signal transduction pathways (Krebs et al., 1988). Purified casein kinase I1 is generally a tetrameric holoenzyme composed of two CY subunits and two P subunits [a2/3,; see Edelman et al. (1987) and Tuazon and Traugh (1990) for reviews]. At low salt concentrations, casein kinase I1 aggregates into much larger, oligomeric, filamentous structures (Glover, 1986; Mamrack, 1989). The /3 subunit has a molecular weight of 24 000-28 000 (Edelman et al., 1987; Tuazon & Traugh, 1990). The function of the P subunit is unknown, although it is presumed to a have a regulatory role in the holoenzyme (Takio et al., 1987; Tuazon & Traugh, 1990). Its amino acid sequence has been determined from bovine lung casein kinase I1 (Takio et al., 1987) and deduced from cDNA clones isolated from Drosophila (Saxena et al., 1987), HeLa cultured cells (Jakobi et al., 1989), and HepG2 cultured cells (Heller-Harrison et al., 1989). The amino acid sequence of the /3 subunit has also been deduced from a human T-cell cDNA clone and is identical with the sequence deduced from the HeLa or HepG2 cDNA clones.2 The a subunit has been identified as the catalytic subunit based upon its ability to bind FSBA or 8-azido-ATP, its sequence identity with other protein kinases, and its persistent (but diminished) catalytic activity in the absence of any p subunit (Hathaway et al., 1981; Cochet & Chambaz, 1983b; Feige et al., 1983; Pyerin et al., 1987; Takio et al., 1987; Saxena et al., 1987; Chen-Wu et al., 1988; Meisner et al., 1989; Tuazon & Traugh, 1990). Often, more than one discrete a-like subunit is present in purified casein kinase 11; the a subunit has a molecular weight of 40000-44000, and the smaller a’ subunit has a molecular weight of 37 000-41 000 (Edelman et al., 1987; Tuazon & Traugh, 1990). In these cases, the tetrameric holoenzyme structure could be exclusively a d & or could be a mixture of two or three possible tetrameric forms: a2P2, and CY CY'/^,. Some evidence has suggested that the CY subunits are related by posttranslational modification, such as by limited proteolysis of the a subunit to generate the smaller a’ subunit (DePaoli-Roach et al., 1981; Hathaway & Traugh, 1982; Hathaway et al., 1983; Meggio & Pinna, 1984; Zandomeni et al., 1988). Other evidence has suggested that CY and a’ subunits are products of different genes (or alternatively spliced transcripts of the same gene). In yeast Saccharomyces cerevisiae and in the bovine, there is immunological and sequence evidence that a and CY’ subunits are distinct polypeptides (Dahmus et al., 1984; Padmanabha & Glover, 1987; Takio et al., 1987; Litchfield et al., 1990). Other casein kinase I1 preparations contain only one type of a subunit (Le., CY^@^), with molecular weight values of the CY subunit varying from 36000 to 42000 (Tuazon & Traugh, 1990). There are even some purified preparations that have many enzymatic characteristics of casein kinase I1 that lack a /3 subunit (Tuazon & Traugh, 1990). Despite having been identified and studied for 15-20 years, many of the molecular properties and cellular functions of casein kinase I1 remain obscure. In an effort to discover more about the properties and functions of mammalian casein kinase 11, we isolated cDNA clones that encode each of the subunits. This report describes the isolation of two related human T-cell

* F. J. Lozeman and E. G . Krebs, unpublished results.

Biochemistry, Vol. 29, No. 36, 1990 8437 cDNA clones that encode the kinase 11.

CY

and

CY’ subunits

of casein

EXPERIMENTAL PROCEDURES Materials. General chemicals and reagents were of analytical grade, obtained from commercial suppliers. T4 DNA ligase and the reagent kit for random primer labeling of DNA were from Boehringer Mannheim. Agarose, 5-bromo-4chloro-3-indolyl 8-D-galactoside (X-gal), isopropyl 1-thio-& D-galactopyranoside (IPTG), nucleic acid molecular size standards, and restriction endonucleases were from Gibco/ BRL or New England Biolabs, Inc. Membranes for blotting or intercepting nucleic acid samples (Le., Nytran, nitrocellulose, and NA-45 DEAE) were from Schleicher and Schuell. 01igo(dT)-cellulose type 7, cesium trifluoroacetate, and Sephadex G-50 Nick columns were from Pharmacia LKB. [CY32P]dATP (3000 Ci/mmol) and [ ( u - ~ ~ S I ~ A(1000 T P Ci/ mmol) were from New England Nuclear. Sequenase enzyme and reagents were obtained from United States Biochemical Corp. M13mp18 and M13mp19 vectors were from Bio-Rad. The plasmid vector pBluescript I1 SK+ was from Stratagene. Guanidine isothiocyanate was from International Biotechnologies, Inc. Other reagents used for recombinant DNA work that are not listed were molecular biology (or equivalent) grade. The Drosophila cDNA clone encoding the a subunit of casein kinase I1 (Xgtl l/dm92; Saxena et al., 1987) was provided by Dr. C. V. C. Glover, University of Georgia. The dm92 insert was subcloned into the EcoRI site in pBluescript I1 SK+ for subsequent manipulations. Oligonucleotides were synthesized with an Applied Biosystems 380A DNA synthesizer (Chemical Synthesis Facility, Howard Hughes Medical Institute, University of Washington). Cultured cell lines were from the American Type Cell Collection. The A43 1 human carcinoma cells were maintained in monolayer cultures in plastic 150-mm culture dishes (Corning) in DME medium containing high (25 mM) glucose and 10% fetal calf serum. HSB-2 lymphoblasts were propagated to plateau density in Eagle’s minimal essential medium with 10% fetal calf serum. Tissues were frozen immediately after collection and stored at -70 OC. Human tissues were obtained from an individual subject immediately after surgery (spleen) or delivery (full-term placenta). Mouse tissues were from male C57 Black 6 hybrid mice. The composition of 20X SSC, 20X SSPE, and 50X Denhardt’s solutions and the deionization and storage of formamide are described in Maniatis et al. (1982). The solution for prehybridization of nitrocellulose membrane plate lifts and genomic Southern blots consisted of 25% formamide, 5X SSPE, 5X Denhardt’s solution, 0.1% SDS, and 100 pg/mL denatured herring sperm DNA. The hybridization solution was the same as the prehybridization solution except for the inclusion of 32P-labeledprobe. The Northern blot prehybridization solution was composed of 50% formamide, 5X SSPE, 5X Denhardt’s solution, 0.1% SDS, and 100 pg/mL denatured herring sperm DNA. The Northern blot hybridization solution consisted of 50% formamide, 5X SSC, 1.5X Denhardt’s solution, 0.1% SDS, 100 pg/mL denatured herring sperm DNA and 32P-labeledprobe. Library Screeningfor a Subunits of Casein Kinase II. The cDNA clones encoding the a subunits of casein kinase I1 were isolated from a human leukemic T-cell library constructed in XgtlO (Littman et al., 1987). The procedure used was essentialy as in Maniatis et al. (1982). C600 Hfl Escherichia coli were infected with approximately 5 X lo5 pfu of the cDNA library and plated onto 150-mm-diameter LB + 0.2% maltose agarose plates (25 000 plaques/plate). After incu-

8438 Biochemistry, Vol. 29, No. 36. 1990 bation, duplicate lifts of each plate were made onto nitrocellulose membranes. The membranes were treated briefly to denature and then neutralize the transferred material, then rinsed (3X SSC) and baked at 80 "C for 2 h. The membranes were prehybridized for 2-4 h at 42 "C and then hybridized at the same temperature for 12-16 h with 1-10 ng/mL SspI fragment of the cDNA clone encoding the a subunit of Drosophila casein kinase I1 (dm92; Saxena et al., 1987). The SspI fragment was 612 bp in length, encompassing most of the coding region of the Drosophila cDNA clone. The SspI fragment was isolated by agarose gel electrophoresis and band interception onto an NA-45 DEAE membrane (intercepted on and eluted from the membrane according to the manufacturer's suggestions) and labeled with [ c Y - ~ * P ] ~ Ato T Pa specific activity of 108-109 dpm/gg using the random primer method (Feinberg & Vogelstein, 1983, 1984). Following hybridization, the membranes were washed 3 times in 2X SSC-O.l% SDS at room temperature for 15 min each time and then 1 h in 1 X SSC-0.1% SDS at 45 "C. Membranes were exposed to Kodak X-Omat AR film with an intensifying screen at -70 "C as required (as long as 72 h). Positive clones were plaque-purified by three additional rounds of screening (Maniatis et al., 1982) using the conditions described above. Subcloning and Sequencing of the cDNAs. The procedures used to prepare, analyze, and subclone the h bacteriophage clones are described by Maniatis et al. (1982, 1989). Briefly, h bacteriophage DNA of the chosen positive clones was isolated from plate lysates, incubation with DNase I and RNase, precipitation with NaCl and poly(ethy1ene glycol) (PEG SOOO), several rounds of extraction with phenol and chloroform, and ethanol precipitation. The cDNA inserts were analyzed by agarose gel electrophoresis after digestion of the crude preparation of h bacteriophage DNA with EcoRI. Various restriction endonuclease fragments of each of the chosen clones were isolated on and eluted from NA-45 membranes (see above) and then ligated into M13 or pBluescript I1 SK' vectors. Aliquots of the ligation reaction mixtures were used to transform E . coli Jm109. Both single-stranded DNA, generated from subclones in M13 vectors, and double-stranded DNA from subclones in SK+ vector were used as templates for dideoxynucleotide sequencing (Sanger et al., 1977), according to the instructions provided by the supplier of the Sequenase enzyme and reagents. Oligonucleotides used as primers in the sequencing reactions either were commercially available or were synthesized from a previously determined sequence (see Materials). There were several regions which were difficult to sequence using the standard protocols; these problems were characterized by pauses and compression in the observed sequence. When pauses were encountered, sequencing was performed only with single-stranded DNA template with higher incubation temperatures for the labeling (37 "C) and termination (50 "C) portions of the reactions. When compression was encountered, the labeling reactions were performed with dITP as suggested by the supplier of the Sequenase enzyme and reagents. The compilation of sequence data was done with Pustell sequence analysis software (International Biotechnologies, Inc.). Comparisons of nucleotide or amino acid sequence to Genbank (Release 61 .O) or Protein Identification Resource (Release 21.0) data banks were accomplished with the Intelligenetics Suite of programs (Intelligenetics, Inc., Mountain View, CA) run on a VAX/VMS computer system. Comparisons were made with the FASTDB program. Genomic Southern Blot Analysis. Human genomic DNA was isolated from venous blood as described by Herrmann and

Lozeman et al. Frischauf (1987). The genomic DNA was digested with restriction endonucleases, electrophoresed on a 1% agarose gel, and transferred to a nitrocellulose membrane (Maniatis et al., 1982). After baking for 2 h at 80 "C, the membrane was prehybridized and hybridized as described above, using specific restriction endonuclease fragments of the human T-cell cDNA clones (see above for the procedures to generate the recombinant plasmids, isolate, and label the fragment). The washing of the membrane was performed as described above, with an extra washing step in 0.2X SSC-0.1% SDS for 30 min at 55 OC. Northern Blot Analysis. Total cellular RNA was extracted from frozen tissue samples or cultured cell samples as described by Okayama et al. (1987). The pieces of frozen tissue samples were disrupted in guanidine isothiocyanate solution with a Polytron homogenizer (Kinetimatica GmbH). A43 1 cells were harvested in phosphate-buffered saline from confluent monolayers, and HSB-2 lymphoblasts were harvested from suspension culture. Cells were collected by centrifugation at 500g for 10 min. Cell pellets were washed twice in several volumes of phosphate-buffered saline by resuspension and centrifugation (as above) before lysis and denaturation in guanidine isothiocyanate. Poly(A)-rich RNA was prepared from the total RNA samples using oligo(dT)-cellulose (Pharmacia) columns according to the manufacturer's suggestions. RNA samples were electrophoresed on a 1% agarose-0.66 M formaldehyde gel and then transferred to Nytran membrane (Fourney et al., 1988). Membranes were baked at 80 "C and then placed in Northern blot prehybridization solution for 2-4 h at 37 "C. Hybridization was performed for 12-16 h at 37 OC in Northern blot hybridization solution using the specific restriction endonuclease fragments of the human T-cell cDNA clones as probes (see above for the labeling procedure). Hybridized Nytran membranes were washed twice in 6X SSC0.1% SDS at room temperature for 15 min each, then twice in 1 X SSC-0.5% SDS at 37 "C for 15 min each, and finally in 0.2X SSC-0.5% SDS at 48-55 "C for 15 min. Sequencing of Peptides from Bovine Lung Casein Kinase II a Subunits. The purification of casein kinase I1 from bovine lung and the subsequent isolation of a subunit peptides is described by Takio et al. (1987). Peptides were generated by cleavage of the a subunits at methionine or lysine residues with cyanogen bromide or Achromobacter protease I. Large fragments were subdigested with TPCK-treated trypsin, Staphylococcus aureus V8 protease, or chymotrypsin. Amino-terminal amino acid sequence analyses were performed by automated Edman degradation as described by Takio et al. (1987). RESULTS Identification and Sequence of cDNA Clones Encoding Casein Kinase I I a Subunits. Upon screening of the human T-cell cDNA library with the fragment of the Drosophila a subunit cDNA (Le., dm92), five positive clones were identified. These clones all remained positive through plaque purification. X bacteriophage DNA was prepared from each of the positive clones. One of the clones had an EcoRI insert which was only about 0.35 kb in length. Two of the clones had EcoRI inserts of about 2.2 kb; the insert of one of these clones (hT 4.1) was subcloned and sequenced. A schematic of the sequencing, the nucleotide sequence, and the deduced amino acid sequence of the clone are presented in Figure 1. Clone hT4.1 contained an open reading frame of 1 176 bp, initiated at an ATG codon starting at nucleotide 157 and terminated at a TGA codon ending at nucleotide 1332. The open reading frame of the cDNA was preceded by 156 bp,

Biochemistry, Vol. 29, No. 36, 1990 8439

Sequence of Casein Kinase I1 a and a' Subunits

A

hT4.1 1

151

2203

1332

3 'UT

5'UT

B 1

111 1

gaattCggAGGGGAGAGCGGCCGCCGCC~GCCGCTTCCACCACAGTTTGAAGAAAACAGGTCTGAAACAAGGTCTTACCCCCAGCTGCTTCTGAACACAGTGACTGCC I CTG AGATCTCCAAACATCAAGTCCAGCTTTGTCCGCCAACCTGTCTGACATGTCGGGACCCGTGCC~GCAGGGCCAGAGTTTACACAGATGTTAATACACACAGACCTCGAG M S G P V P S R A R V Y T D V N T H R P R E

221 23

AATACTGGGATTACGAGTCACATGTGGTGGAATGGG~TCAAGATGACTACCAGCTGGTTCG~TTAGGCCGAGGTAAATACAGTGAAGTATTTGAAGCCATCAAC Y W D Y E S H V V E W G N Q D D Y Q L V R K L G R G K Y S E V F E A I N

33 1 59

ATCACAAATAATGAAAAAGTTGTTGTTAAAATTCTCAAGCCAGT~GAAG~TTAAGCGTGAAATAAAGATTTTGGAGAATTTGAGAGGAGGTCCCAACATCAT

441 96

CACACTGGCAGACATTGTAAAAGACCCTGTGTGTCACGAACCCCCGCCTTGGTTTTTG~CACGT~CAACACAGACTTCAAGCAATTGTACCAGACGTTAACAGACTATG T L A D I V K D P V S R T P A L V F E H V N N T D F K Q L Y Q T L T D Y D

55 1 133

ATATTCGATTTTACATGTATGAGATTCTGAAGGCCCTGGATTATTGTCACAGCATGGGAATTATGCACAGAGATGTCAAGCCCCATAATGTCATGATTGATCATGAGCAC I R F Y M Y E I L K A L D Y C . H S M G I M H R D V K P H N V M 1 D H E H

661 169

AGAAAGCTACGACTAATAGACTGGGGTTTGGCTGAGTTTTATCATCCTGGCCAAGAATATAATGTCCGAGTTGCTTCCCGATACTTCAAAGGTCCTGAGCTACTTGTAGA R K L R L I D W G L A E F Y H P G Q E Y N V R V A S R Y F K G P E L L V D

771 206

CTATCAGATGTACGATTATAGTTTGGATATGTGGAGTTTGGGTTGTATGCTGGCAAGTATGATCTTTCGGAA~AGCCATTTTTCCATGGACATGACAATTATGATCAGT Y Q M Y D Y S L D M W S L G C M L A S M I F R K E P F F H G H D N Y D Q L

881 243

TGGTGAGGATAGCCAAGGTTCTGGGGACAGAAGATTTATATGACTATATTGAC~TAC~CATTGAATTAGATCCACGTTTCAATGATATCTTGGGCAGACACTCTCGA V R I A K V L G T E D L Y D Y I D K Y N I E L D P R F N D I L G R H S R

991 279

AAGCGATGGGAACGCTTTGTCCACAGTGAAAATCAGCACCTTGTCAGCCCTGAGGCCTTGGATTTCCTGGACAAACTGCTGCGATATGACCACCAGTCACGGCTTACTGC K R W E R F V H S E N Q H L V S P E A L D F L D K L L R Y D H Q S R L T A

I

T

N

N

E

K

V

V

V

K

I

L

K

p

V

K

K

K

.

K

I

K

R

E

I

K

I

L

E

N

L

R

G

G

P

N

I

1

1101 AAGAGAGGCAATGGAGCACCCCTATTTCTACACTGTTGTGAAGGACCAGGCTCGAATGGGTTCATCTAGCATGCCAGGGGGCAGTACGCCCGTCAGCAGCGCC~TATGA 31 6 R E A M E H P Y F Y T V V K D Q A R M G S S S M P G G S T P V S S A N M M 1211 353

TGTCAGGGATTTCTTCAGTGCAACCCCTTCACCCCTTGGACCTCTGGCA~CTCACCAGTGATTGCT~TGCCAACCCCCTTG~ATGCCTGTTCCAGCTGCCGCTGGC S G I S S V P T P S P L G P L A G S P V I A A A N P L G M P V P A A A G

1321 GCTCAGCAGTAACGGCCCTATCTGTCTCCTGATGCCTGAGCAGAGGTGGGGGAGTCCACCCTCTCCTTGATGCAGCTTGCGCCTGGCGGGGA~GGTGAAACACTTCAGA 38 9 A Q Q Ter 1431

AGCACCGTGTCTGAACCGTTGCTTGTGGATTTATAGTAGTTCAGTCA V T A T A A T A G G C T G A T T T T C T T T T T T C T T T T TVT T T T T T A A C T C G A A C T T T T C A

1541

TAACTCAGGGGATTCCCTGAAAAATTACCTGCAGGTGGAATATTTCATGGAC~TTTTTTTTTCTCCCCTCCCAAATTTAGTTCCTCATCACAAAAGAACAAAGATAAA

1651

CCAGCCTCAATCCCGGCTGCTGCATTTAGGTGGAGACTTCTTCCCATTCCCACCATTGTTCCTCCACCGTCCCACACTTTAGGGGGTTGGTATCTCGTGCTCTTCTCCAG

1761

AGATTACAAAAATGTAGCTTCTCAGGGGAGGCAGGAAGAAAGGAAGGAAGGAAAGAAGGAAGGGAGGACCCAATCTATAGGAGCAGTGGACTGCTTGCTGGTCGCTTACA

1871 1981

AAAAvCT~CTGTTCAGT~G~TCATGTCTTACTGATCTAACCCTAAATCCAACTCATTTATACTTTTATTTTTAGTTCAGTTTATGTTGATACCTTCCCTCCCAGGC V

2091

TCCTTACCTTGGTCTTTTCCCTGTTCATCTCCCAACATGCTGTGCTCCATAGCTGGTAGGAGAGGGAAGGC~TCTTTCTTAGTTTTCTTT~TCT~GGCCATTTTGAA A I TTC

2201

1: Human casein kinase I1 a subunit: structure, sequencing strategy, nucleotide sequence, and amino acid sequence of the cDNA clone hT4.1. (A) Structure of the cDNA clone hT4.1 showing several restriction endonuclease sites and the 5' untranslated region (YUT), the coding region (the shaded area), and the 3' untranslated region (3'UT). The arrows under the diagram represent the direction and amount of sequence data determined from individual sequencing reactions by the dideoxy chain termination method. (B) Nucleotide sequence of the hT4.1 cDNA clone is presented with the amino acid sequence deduced from the coding region. The amino acids are depicted by the single-letter abbreviation. The XgtlO cDNA library was constructed with EcoRI linkers (Littman et al., 1987). Lowercase letters at the 5' end of the clone indicate an EcoRI linker. The 3' end of the clone did not appear to have an EcoRI linker, so the EcoRI site is not in lowercase letters. "Ter" indicates the termination codon at the end of the coding region. The cDNA clone hT4.1 was nearly identical in size and sequence with a cDNA clone isolated from HepG2 cells (Meisner et al., 1989). Differences between the two sequences are indicated as follows: nucleotides that are different in the HepG2 sequence are given below the cprresponding underlined nucleotide in the hT4.1 sequence; nucleotide bases that are above the (V) symbol are not in the HepG2 sequence; A, additional A nucleotide in the HepG2 sequence; the underlined nucleotide above 7 corresponds to nucleotide 7 in the HepG2 sequence; the underlined nucleotide above (I) is the last nucleotide in the HepG2 sequence. FIGURE

which contained no initiation codons. The open reading frame was followed by a 3' untranslated end of 871 bp. Neither a polyadenylated tail nor a polyadenylation signal (AATAA; Proudfoot & Brownlee, 1976) was present in the 3' untranslated end of the cDNA, and the EcoRI site at the 3' end of the clone did not appear to be derived from the EcoRI linker used to construct the cDNA library (Figure 1). The lack of a polyadenylation signal, and the presence of an EcoRI site probably present in the mRNA, suggested that the cDNA

insert was not a full-length representative of the mRNA. The size and nucleotide sequence of hT4.1 were virtually identical with a cDNA clone from a HepG2 library which was recently reported (Meisner et al., 1989). The minor discrepancies between the two clones were substitutions, insertions, or deletions of single nucleotides, present in the 5' or 3' noncoding regions of the clones (see Figure 1). These clones encoded an identical polypeptide which was undoubtedly the a subunit of casein kinase 11. This conclusion could be made because

8440 Biochemistry, Vol. 29, No. 36, 1990 the partial amino acid sequence derived specifically from the a subunit (i.e., M,45000) of bovine testis casein kinase I1 (Litchfield et al., 1990) was identical with the sequence encoded by the human hT4.1 or HepG2 cDNA clones (see Discussion). The remaining two of the five isolated clones each had two EcoRI inserts of similar size, one insert being about 0.5 kb and the other 1.25-1.35 kb. The insert of one of the clones (hT4.2) was subcloned and partially sequenced. This clone was I752 bp in length, with an open reading frame of 984 bp that was 74% identical with the nucleotide sequence of the corresponding region of the hT4.1 clone (data not shown). However, this clone appeared to be incomplete, since the open reading frame encoding the polypeptide similar to the a subunit was not preceded by a 5’ untranslated region or an initiation codon. A 479 bp EcoRI fragment of the hT4.2 clone was used to screen an additional 8 X lo5 individual plaques of the human T-cell cDNA library according to the same procedures used initially with the Drosophila cDNA probe (see Experimental Procedures). Seven positive clones were identified and isolated as a result of the screen with the fragment of hT4.2. Restriction endonuclease analysis of the h bacteriophage DNA from each of these isolated clones indicated three of the seven were similar to clone hT4.2, but longer in the 5’ region of the clones. One of these clones (hT9.1) was subcloned in various pieces into M13 and pBluescript I1 SK’ vectors and sequenced. A schematic of the sequencing, the nucleotide sequence, and the deduced amino acid sequence of clone hT9.1 are presented in Figure 2. The cDNA clone hT9.1 was 1774 bp in length with a 5’ untranslated region of 17 1 nucleotides. The ATG initiation codon starting at nucleotide 172 was the first in the clone, and the sequence immediately preceding the ATG corresponded to the consensus eukaryotic initiator sequence (CCtCCAUG) as defined by Kozak (1984). The initiation codon was followed by an open reading frame of 1053 bp. The 3’ untranslated region contained a consensus polyadenylation signal, AATAAA (Proudfoot & Brownlee, 1976), which was 12 bp upstream from the poly(A) tail. It was not possible to sequence the poly(A) tail and surrounding sequence on both strands. The DNA chains synthesized in vitro that contained the poly(A) region [or the poly(T) region of the complement strand] varied in length by one to three bases, and thus made any sequence that followed this region unreadable. This problem may have been caused by heterologous template [Le., template varying in length by one or more nucleotides in the poly(A) region] generated when template samples were propagated in E . coli. Alternatively, the inability to sequence through this region may have been due to stuttering of the polymerase within the poly(A) region during the sequencing reaction. As indicated in Figure 2, the length of the poly(A) tail of the clone was estimated at 80-82 bp. It was obvious from the nucleotide sequence that the cDNA clones hT4.1 and hT9.1 represent very related, but distinct, mRNA transcripts. The nucleotide sequence of hT9.1 was 76% identical with the aligned sequence of hT4.1 between bp 168 and 1 170 of hT4.1. There was little homology between the two clones outside of these regions. The partial amino acid sequence determined specifically from the a’ subunit (Le., M , 40000) of bovine testis casein kinase I1 (Litchfield et al., 1990) was identical with the sequence encoded by hT9.1. Therefore, clone hT9.1 encodes the a’ subunit of casein kinase I1 (see Discussion). Amino Acid Sequence of Casein Kinase I1 a Subunits. The predicted molecular weights of the hT4.1- and hT9.1-encoded

Lozeman et al. polypeptides were 45 160 and 41 400, respectively. An alignment of these polypeptides is presented in Figure 3. As seen in Figure 3, most of the differences between the two polypeptides were at the carboxy terminus, with a few positions of heterogeneity scattered throughout the remainder of the polypeptides. The first 330 amino acids of the polypeptide encoded by hT9.1 were 88% identical with the aligned hT4.1-encoded polypeptide and 8 1% identical with the Drosophila a subunit [not shown; see Saxena et al. (1987)]. A considerable amount of amino acid sequence was determined from a mixture of the a subunits of bovine lung casein kinase 11; a small portion of this sequence has been reported previously (Takio et al., 1987). After separation of the a subunits from the /3 subunit by size-exclusion chromatography, various proteolytic fragments of the a subunits were generated by CNBr or enzymatic digestion and isolated by a combination of size-exclusion and reverse-phase chromatography. Of the sequence determined by Edman degradation of the various proteolytic fragments, 3 14 residues could be best aligned with hT4.1-encoded polypeptide (i.e., the human a subunit; Figure 3), and 221 residues could be best aligned with the hT9.1encoded polypeptide (i.e., the human a’ subunit; Figure 3). Only two discrepancies were observed in the alignments despite comparing the sequence from bovine peptides with the sequence deduced from human cDNA clones. These discrepancies are discussed in the next paragraph. The amino terminus of one of the isolated peptides appeared to be blocked, since it was not susceptible to Edman degradation. This peptide was sequenced after mild acid treatment that promoted an acyl N to 0 shift (Takio et al., 1987), exposing the SGPVPSR sequence (Figure 3). Amino acid analysis verified the lack of additional amino acids in this peptide. Thus, it appears that the initiator methionine was removed in vivo. There were 29 positions of heterogeneity identified in the mixture of bovine lung sequences after they were aligned with the polypeptides deduced from the human cDNA clones. All but one of these positions of heterogeneity was also observed in the deduced cDNA polypeptides (Figure 3). The exception was at residue 252 (see numbering in Figure 3) where both clones coded for glutamic acid but a bovine peptide corresponding to the a’ subunit sequence displayed aspartic acid. This suggests a conservative amino acid change in the a’ subunit of human and bovine casein kinase I1 at this position. Another conservative amino acid change between bovine and human a’subunits appeared at position 229 (Figure 3). Lysine was present in the human a subunit (Le., encoded by hT4.1) while the human a’subunit (encoded by hT9.1) contained an arginine at this position. Bovine peptides corresponding to either the cy or the a’ subunit contained only lysine at this position (Figure 3). Although alone these data did not show which of the cDNA clones encoded the a subunit and which encoded the a’ subunit, the nearly perfect identity with the partial bovine lung sequences did show that hT4.1 and hT9.1 encoded bona fide a and a‘ subunits of casein kinase 11. Nucleic Acid and Protein Sequence Data Bank Comparisons. Comparisons of the nucleotide or deduced amino acid sequences of hT4.1 and hT9.1 were made to recent releases of the GenBank nucleic acid data bank or the Protein Identification Resource protein data bank. As expected, homologies of hT4.1 and hT9.1 nucleotide or amino acid sequences were with the casein kinase I1 a subunit clones from HepG2 cells (Meisner et al., 1989), Drosophila (Saxena et al., 1987), and S. cereoisiae CKAl gene (Chen-Wu et al., 1988). Other proteins or DNA clones identified in the search showed much

Sequence of Casein Kinase 11 a and a’ Subunits

Biochemistry, Vol. 29, No. 36, I990

8441

A k

k

h

d

3

u

w hT9.1 1

172

1224

1774

5 ‘UT

3 ‘UT

B 1

gaattcggTGTCACCCAGGCTGGAGTGCAGTGGCGCAATCTCAGCTCACTGCAACCTCCACCTCCCTGGTTCAAGCGATTCTCCTGCCTCCTCCGCCCGACGCCCCGCGT

111 1

CCCCCGCCGCGCCGCCGCCGCCACCCTCTGCGCCCCGCGCCGCCCCCCGGTCCCGCCCGCCATGCCCGGCCCGGCCGCGGGCAGCAGGGCCCGGGTCTACGCCGAGGTGA M P G P A A G S R A R V Y A E V N

221 18

ACAGTCTGAGGAGCCGCGAGTACTGGGACTACGAGGCTCACGTCCCGAGCTGGGGTATCAAGATGATTACCAACTGGTTCGAAAACTTGGTCGGGGAAAATATAGTGAA S L R S R E Y W D Y E A H V P S W G N N D D Y Q L V R K L G R G K Y S E

331 54

GTATTTGAGGCCATTAATATCACCACATGAGAGAGTGGTTGT~TCCTGAAGCCAGTGAAG~GAAGATAAAACGAGAGGTTAAGATTCTGGAGAACCTTCG V F E A I N I T N N I R V V V K I L K P V K K K K I K R E V K I L E N L R

441 91

TGGTGGAACATATCATTAGCTGATTGACACTGTAGGACCCCGTGTCAGACACCAGCTTTGGTATTTGAATATATCAATAATACAGATTTTAAGCAACTCTACC G G T N I I K L I D T V K D P V S K T P A L V F E Y I N N T D F K Q L Y Q

551 128

AGATCCTGACAGACTTTGATATCCGGTTTTATATGTATGAACTACTTAGCTCTGGATTACTGCCACAGCAAGGGAATCATGCACAGGGATGTGACCTCACAATGTC I L T D F D I R F Y M Y E L L K A L D Y C H S K G I M H R D V K P H N V

661 164

M

771 201

ACCAGAGCTCCTCGTGGACTATCAGATGTATGATTATAGCTTGGACATGTGGAGTTTGGGCTGTATGTTAGCAAGCATGATCTTTCGAAGGGAACCATTCTTCCATGGAC P E L L V D Y Q M Y D Y S L D M W S L G C M L A S M I F R R E P F F H G Q

881 238

AGGACAACTATGACCAGCTTGTTCGCATTGCCAAGGTTCTGGGTACAGAAGAACTGTATGGGTATCTGAAGAAGTATCACATAGACCTAGATCCACACTTCAACGATATC D N Y D Q L V R I A K V L G T E E L Y G Y L K K Y G I D L D P H F N D I

991 274

CTGGGACAACATTCACGGACGCTGGGAAAACTTTATCCATAGTGAGAACAGACACCTTGTCAGCCCTGAGGCCCTAGATCTTCTGGACACTTCTGCGATACGACCA L G Q H S R K R W E N F I H S E N R H L V S P E A L D L L D K L L R Y D H

ATGATAGATCACCAACAGAGCTGCGACTGATAGATTGGGGTCTGGCAGAATTCTATCATCCTGCTCAGGAGTACAATGTTCGTGTAGCCTCAAGGTACTTCAAGGG

I

D

H

Q

Q

K

K

L

R

L

I

D

W

G

L

A

E

F

Y

H

P

A

Q

E

Y

N

V

R

V

A

S

R

Y

F

L

G

1101 311

TCAACAGAGACTGACTGCCAGAGGCCATGGAGCACCCATACTTCTACCCTGTGGTGAAGGAGCAGTCCCAGCCTTGTGCAGAC~TGCTGTGCTTTCCAGTGGTCTCA Q Q R L T A K E A M E H P Y F Y P V V K E Q S Q P C A D N A V L S S G L T

1211 348

CGGCAGCACGATGAGACTGGAGCGACGGGTCTGTTGCGGTTCTCCCACTTTTCCATAAGCAGAACAAGAACCATCACGTCTTAACGCGTATAGAGAGATCACG A A R Ter

1321

TTCCGTGAGCAGACACAAAACGGTGGCAGGTTTGGCGAGCACGAACTAGACCAAGCGAAGGGCAGCCCACCACCGTATATCACCTCACTTCCGAATGT~GGCTCA

1431

C T T G C C T T T G G C T T C C T G T T G A C T T C T T C C C G A C C C A G A A C T C G T C

1541

CCGTGGCCGCCTGTTTTTCCAGCACCACGCTAACTAGCTGACCACAGACTCCACAGTGGGGGGACGGGCGCAGTATGTGGCATGGCGGCAGTTACATATTATTATTTT

1651

A

1761

AAAAAAccgaattc

A

A

A

G

T

A

T

A

T

A

T

T

A

T

T

G

D

G

G

T

T

T

T

A

A

A

A

G

M

FIGURE2: Human casein kinase I1 a’ subunit: structure, sequencing strategy, nucleotide sequence, and amino acid sequence of the cDNA

clone hT9.1. (A) Structure of the cDNA clone hT9.1 showing several restriction endonuclease sites and the 5’ untranslated region (S’UT), the coding region (the shaded area), and the 3’ untranslated region (3’UT). The arrows under the diagram represent the direction and amount of sequence data determined from individual sequencing reactions by the dideoxy chain termination method. It was not possible to determine the sequence 3’ of the poly(A) region regardless of the strand used as the template in the reactions. Therefore, the sequence on either side of the poly(A) region could not be confirmed by sequence data from the complement strand. It was estimated that the poly(A) region was 80-82 nucleotides in length. (B) Nucleotide sequence of the hT9.1 cDNA clone is presented with the amino acid sequence deduced from the coding region. The amino acids are depicted by the single-letter abbreviation. The XgtlO library was constructed with EcoRI linkers (Littman et al., 1987). Lowercase letters at the 5’ and 3’ ends of the clone indicate EcoRI linkers. “Ter” indicates the termination codon at the end of the coding region. The consensus polyadenylation signal sequence (Proudfoot & Brownlee, 1976) is underlined.

less identity. They included amino acid or DNA sequences of other protein kinases and, in particular, the protein encoded by the cell division control gene CDC28 in S . cerevisiae (Lorincz & Reed, 1984) or its homologues in Schizosaccharomyces pombe (cdc2; Hindley & Phear, 1984) and human (cdc2-hs; Lee & Nurse, 1987). The homology of casein kinase I1 a subunit with CDC28 and cdc2 proteins was noted previously by Takio et al. (1987). Detection of a Subunit mRNA in Various Tissues and Cultured Cell Lines. Northern blot analysis was performed

on poly(A)-rich RNA isolated from two different human cultured cell lines and from three mouse tissues using the hT4.1 ( a subunit) and hT9.1 (a’ subunit) cDNA clones (or portions of the clones) as probes. In mouse brain, thymus, and testis samples, the hT4.1 clone hybridized to R N A with sizes of about 4.6 and 3.1 kb (Figure 4A); a band of about 1.8 kb was visible with a longer exposure or if more poly(A)-rich RNA (20 pg) was analyzed (data not shown). In RNA from the human cultured cell lines, bands of about 4.5, 2.8, and 1.7 kb were detected (Figure 4B,C). The same results were obtained

8442 Biochemistry, Vol. 29, No. 36, 1990

Lozeman et al.

Human a subunit from cDNA hT4.1: Human a ’ subunit from cDNA hT9.1:

10 20 30 MSGPVPSRARVYTDVNTHRPREYWDYESHV MPGPAAG******AE**SL*S*******Af*

110 120 130 140 150 160 170 VKDPVSRTPALVFEHVNNTDFKQLYQTLTDYDIRFYMYEILKALDYCHSMGI~RDVKPHN~IDHEHRK

-.

******K*******YI**********I***F****L*********K****************QQK*

__

180 190 200 210 220 230 240 LRLIDWGLAEFYHPGQEYNVRVASRYFKGPELLVYQMYDYSLDMWSLGCMLASMIFRKEPFFHGHDNYD **************A****************************************

250 2 60 270 280 2 90 300 310 QLVRIAKVLGTEDLYDYIDKYNIELDPRFNDILGRHSRKRWERFVHSENQHLVSPEALDFLDKLLRYDHQ **G*LK**H*D***H******Q*****N*I**N*I****R*********L**********

320

330

340

350

360

370

380

S R L T A R E A M E H P Y F Y T W K D Q A R M G S S S M P G G S T P V S S A N M P L G

Q****K*********P***E*SQPCADNAVLSSGLTAARtga 390 MPVPAAAGAQQtga FIGURE 3: Alignment of the polypeptides deduced from the human cDNA clones hT4.1 and hT9.1 with segments of amino acid sequences determined from bovine lung casein kinase 11. The amino acid sequences (one-letter code) deduced from the two cDNA clones were aligned without gaps; an asterisk indicates that the amino acid residue coded by hT9.1 (a’subunit) was the same as for hT4.1 ( a subunit). The end of each polypeptide encoded by the respective cDNA clone is indicated by “tga”. The numbering above the sequence relates to the hT4.1-encoded amino acid sequence. Note that this numbering is also used when referring to the hT9.1-encoded sequence. Underlined amino acid sequence represents residues observed by automated Edman degradation of various peptides from the mixture of the a and a’ subunits of bovine lung casein kinase I 1 (see Experimental Procedures). Some bovine lung sequences aligned best to the hT4.1-encoded polypeptide (shown by underlining of the first row of sequence); others aligned best to the hT9.1-encoded polypeptide (shown by underlining of the second row of sequence). Two differences between the bovine lung sequences and the sequence deduced from the hT9.1 clone are shown by circled letters at residues 229

and 252.

with poly(A)-rich RNA (5-10 pg) from human spleen or from full-term human placenta with the hT4.1 cDNA clone as the probe (data not shown), indicating that the cultured human cells express the same RNA transcripts as at least two different tissues. The smallest RNA (1.7-1.8 kb) had the least prominent signal in each of the samples, suggesting that it was present in the lowest concentration or that it had the lowest homology with the hT4.1 cDNA clone. The 2.8-kb RNA was more prominent than the 4.5-kb RNA in HSB-2 cells, but in the other samples, those bands were about equal in intensity. Multiple hybridizing bands were also evident by Northern blot analysis when a portion of the hT9.1 cDNA clone was used as the probe. Bands of about 5.6, 3.2, and 2.1 kb were observed in the three mouse tissues, the 2.1-kb band being the most intense (Figure 4A). The hT9.1 clone hybridized to RNA of 5.7, 4.5, 2.8, 2.0, and 1.0 kb in the A431 sample (Figure 4B) and 5.7, 3.0, 2.0, and 1.6 kb in the HSB-2 sample (Figure 4C). In the A431 sample, the 4.5- and 2.8-kb bands were present with hybridization with either cDNA clone (Figure 4B). This suggests either that A431 cells contain independent 4.5- and 2.8-kb RNAs both for a and for a’ subunits or that the 4.5- and 2.8-kb RNAs can be recognized by either cDNA clone (Le., the cDNA clones cross-react with the 4.5- and 2.8-kb RNAs in A431 cells). The former explanation seems more plausible, since cross-reaction of the two cDNA clones was not obvious in any of the mouse tissues or in the HSB-2 cells. The amount of RNA recognized by the hT9.1 clone was greater in mouse testis than in thymus or brain (Figure 4A),

especially for the 2.1-kb transcript. This was not true of the transcripts recognized by the hT4.1 clone (Figure 4A). When the same blots were hybridized with a cDNA clone encoding the p subunit of casein kinase 11, the concentration of the major transcript was also greatest in testis, followed by thymus and then brain (data not shown). This is in general agreement with studies on the enzymatic activity of casein kinase 11, since its activity in extracts of testis is greater than in other tissues (Nakajo et al., 1986; Krebs et al., 1988). Analysis of the a Subunits of Casein Kinase II by Genomic Southern Blot. The entire hT4.1 (asubunit) cDNA clone and a ScaI-BglII fragment of the hT9.1 (a’subunit) clone (nucleotides 242-1068) were labeled with [ C X - ~ ~ P I ~and A Tused P to probe a Southern blot of human genomic DNA which had been digested with various restriction endonucleases. The analysis was performed separately on the genomic DNA from two different individuals. The results of the analysis of the DNA from one of the individuals are presented in Figure 5. The results with the DNA from the other individual were identical. There were no EcoRI, HindIII, or BamHI sites within the hT4.1 cDNA clone (see Figure l), but there were at least two hybridizing bands in the lanes of genomic DNA digested with these restriction endonucleases (lanes 1, 4, and 5; Figure sa). There were single sites for PstI and BglII in the hT4.1 cDNA, and, as expected, there was more than one hybridizing band in each of the lanes (lanes 2 and 3, Figure 5a). The autoradiogram of the same blot (washed stringently to remove previous probe) hybridized with the hT9.1 fragment also indicated at least two hybridizing bands in each lane

Sequence of Casein Kinase I1 a and a’ Subunits A

C

B

Biochemistry, Vol. 29, No. 36, 1990 8443

1 2 3 4 5

1 2 3 ‘ kb

A431

-23.1-

a b

9.5

-

7.54.64.4-

1.9-

-

m

2.4

-*

1.3

-

*

-

- 4.4-

-

- 2.32.0-

85 7.5-

(*

4.4 0..

2A1.3-

0.24

-

9.46.6-

0.24

-

0.6-

-

FIGURE 4: Northern blot analysis of mouse and human RNA with the cDNA clones hT4.1 and hT9.1. Poly(A)-rich RNA was separated by agarose-formaldehyde gel electrophoresis and transferred to charged nylon membrane. The membranes were then prehybridized, hybridized in the presence of the appropriate 32P-labeled cDNA fragment, and washed (seeExperimental Procedures). Kodak X-Omat AR film was exposed to washed membranes with an intensifying screen at -70 O C for 5 days. (a) Membrane blots were hybridized with the entire hT4.1 (human a subunit) cDNA clone (EcoRI fragment, 2197 nucleotides). (b) Membrane blots were hybridized with a ScuI-SstI (538 nucleotides) fragment of the hT9.1 (human a’subunit) cDNA clone. (A) Ten micrograms of poly(A)-rich RNA from mouse brain (Br), mouse thymus (Th), and mouse testis (Te). (B) Ten micrograms of poly(A)-rich RNA from A43 I cultured cells. (C)Five micrograms of poly(A)-rich RNA from HSB-2 cultured cells. The positions of RNA molecular size standards (see Experimental Procedures; kb) or mouse 28s and 18s rRNA (4.6 and 1.9 kb) are indicated beside the sample lanes.

(Figure 5b). There were no PstI, BglII, HindIII, or BamHI sites within the fragment of the hT9.1 cDNA clone used, but there were two or more hybridizing bands in each of the lanes of genomic DNA digested with these restriction endonucleases (lanes 2-5, Figure 5b; the very large -23-kb band was not very evident in this autoradiogram but was more evident in the autoradiogram from the other sample of genomic DNA). There was an EcoRI site within the fragment of hT9.1, and bands of -2.5, 8.5, and 23 kb were evident in the lane containing EcoRI-digested genomic DNA (lane 1, Figure 5b). The hT9.1 fragment did not recognize the same sizes of digested genomic DNA as the hT4.1 cDNA clone, indicating that there was little cross-hybridization of the two clones under the conditions used for the analysis. (Bands of equal size were faintly evident in some cases.) It was obvious from the autoradiograms that the two cDNA clones recognized different genomic DNA restriction endonuclease fragments, indicating that the cDNA clones represent two different gene products. The analysis was consistent with a single copy of a gene corresponding to each of the two cDNA clones, each containing one or more introns (with the pertinent restriction endonuclease sites) to explain the multiple and large hybridizing bands. An alternative but more unlikely explanation of the multiple hybridizing bands is that there were additional copies of each gene which were highly related to the respective cDNA clones.

DISCUSSION Two distinct clones were isolated from a human T-cell cDNA library using a fragment of a Drosophila cDNA clone

FIGURE 5: Southern blot analysis of human genomic DNA with the cDNA clones hT4.1 and hT9.1. Genomic DNA from human whole blood was prepared, digested, electrophoresed, and blotted to nitrocellulose membrane as described under Experimental Procedures. The conditions for prehybridization, hybridization, and washing of the membrane blot are also given under Experimental Procedures. Kodak X-Omat AR film was exposed to the washed membrane blot with an intensifying screen at -70 O C for 5 days. Ten micrograms of genomic DNA was used for each restriction enzyme digest: lane 1, EcoRI; lane 2, Psrl; lane 3, BglII; lane 4, HindIII; lane 5, BumHI. The membrane blot was hybridized either with the entire hT4.1 cDNA clone (a) or with a ScuI-BglII (827 nucleotides) fragment of the hT9.1 cDNA clone (b). The same nitrocellulose membrane blot was used for analysis with each of the cDNA probes. Hybridized probe was removed by a stringent wash (twice, 15 min each in 0.2X SSC and 0.1% SDS at 95 “C). Removal of the probe was verified by autoradiography for 3 days. The positions of DNA standard size markers (kb) are indicated beside the samples lanes.

encoding the a subunit of casein kinase 11. These two cDNA clones encode distinct, but homologous, polypeptides. The sequences of the polypeptides encoded by the two human cDNA clones are nearly identical with the partial amino acid sequences determined from a mixture of a and a’ subunits from purified bovine lung casein kinase I1 (Figure 3). The isolation of the two human cDNA clones confirms previous immunological and sequence data, which suggested that the a and d subunits of bovine casein kinase I1 were separate gene products, or products of alternatively spliced transcripts (Dahmus et al., 1984; Takio et al., 1987; Litchfield et al., 1990). The analysis of genomic DNA with the two cDNA clones (Figure 5 ) indicates that the a and a’subunits of human casein kinase I1 are products of distinct genes. From the data presented in this report, it was not possible to conclude which of the cDNA clones encodes the a subunit and which encodes the a’subunit. This was accomplished by comparing the sequences deduced from the human cDNA clones to sequences determined exclusively from the a or the a’ subunit of bovine testis casein kinase I1 (Litchfield et al., 1990). Protein microsequencing techniques were used to generate the specific amino acid sequences from the a and the a’ subunits. Among others, the sequences VLGTEDLYDYIDK and YNIELDPRFNDILGRHSRK were determined

8444 Biochemistry, Vol. 29, No. 36, 1990 from the bovine a subunit. These sequences are identical with residues 248-260 and 26 1-279 in the hT4.1-encoded polypeptide (Figure 3), confirming that hT4.1 encodes the a subunit. The sequences VLGTDELYGYLK and YHIDLDPHFNDILGQHSRK were determined from the bovine a’ subunit. This confirms that the hT9.1 clone encodes the a’ subunit of casein kinase TI, since these peptide sequences are identical (except for one conservative change at residue 252) with residues 248-259 and 261-279 in the hT9.1-encoded polypeptide (see numbering in Figure 3). The conclusions based upon the partial bovine sequence are supported by immunoblot studies with antiserum against a peptide with a sequence EDLY DY IDKYNIEL, corresponding to residues 252-265 i n the hT4. I-encoded polypeptide; this antipeptide antiserum specifically recognizes the a subunit of bovine testis casein kinase I1 (Litchfield et al., 1990). Despite the evidence that a and a’ are distinct subunits, the proteolytic sensitivity of cy and a’ often makes it difficult to distinguish one from another on SDS-polyacrylamide gel electrophoresis. In the purification of bovine testis casein kinase 11, limited proteolysis of the a subunit generates a polypeptide which is difficult to distinguish from the legitimate a’ ~ u b u n i t . ~Therefore, caution should be exercised when interpreting the subunit composition of purified casein kinase I1 by SDS-polyacrylamide gel electrophoresis. The a subunit is the catalytic subunit of casein kinase 11, based upon enzymatic activity in the absence of the P subunit, affinity labeling with ATP analogues, and sequence features common to other protein kinases (see references in the introduction). The a’ subunit also has the sequence and primary structural features common to other protein kinases. In fact, all of the sequence features of protein kinases that have been compiled by Hanks et al. (1988) are conserved between the cy and cy’ subunits. The basic region that follows the critical nucleotide binding lysine (KPVKKKKIKR; residues 71-80 in the hT4.1-encoded polypeptide) is a sequence feature in casein kinase I1 a subunits from numerous organisms (Padmanabha et a]., 1990) including the human a and cy’ subunits (Figure 3). This region could be responsible for interaction with the substrates of casein kinase I1 [see Tuazon and Traugh (1990)], considering that the major substrate specificity requirement is for acidic residues in the first few positions (and especially at the +3 position) immediately following the phosphorylated serine or threonine (Meggio et al., 1984; Marin et al., 1986; Kuenzel et al., 1987). This basic region may also be involved in the interaction with the /3 subunit. The sequence DLEPDEELED (residues 51-60 of the bovine lung (3 subunit; Takio et al., 1987) contains seven acidic residues and is highly conserved between organisms (Saxena et a]., 1987; Takio et ai., 1987; Jakobi et al., 1989; Heller-Harrison et ai., 1989). Thus, it is possible that an interaction between this acidic region in the @ subunit and the basic region in the a and a’ subunits is partly responsible for the interaction of the subunits within the holoenzyme. However, it is unlikely that this is the only motif responsible for the subunit interactions since high salt concentration is not sufficient to dissociate the holoenzyme (Meggio & Pinna, 1984). The effects of polyanionic and polycationic compounds on the activity and the autophosphorylation of casein kinase I1 (Maenpaa, 1977; Hathaway & Traugh, 1982; Meggio & Pinna, 1984; Mamrack, 1989) are consistent with these proposed roles for the highly charged motifs KPVKKKKTKR and DLEPDEELED in D. W. Litchfield and E. G. Krebs, unpublished results.

Lozeman et al. substrate recognition, activity, and subunit interaction. Recent studies have suggested another intriguing function for the basic KPVKKKKIKR region in the a and a’ subunits. Several investigators have noted that nuclear proteins contain specific sequence motifs which direct the protein to the nucleus. The motifs are short segments of basic amino acids (Kalderon et al., 1984; Richardson et al., 1986; Lyons et al., 1987; Dang & Lee, 1989). Dang and Lee (1989) determined that a helix-breaking residue (proline or glycine) followed by a t least three basic residues in the next seven to nine carboxy-terminal residues was an effective nuclear targeting sequence. The KPVKKKKIKR sequence certainly falls into this description, implying that the a and cy’ subunits could be targeted to the nucleus by this motif. Casein kinase I1 is found in the nucleus where it may consist of both an a and an a’ subunit (Thornburg & Lindell, 1977; Thornburg et al., 1979; Inoue et al., 1984; Qi et al., 1986). However, others report that nuclear casein kinase I1 contains only a single a subunit (Rose et al., 1981; Goueli et al., 1986). The molecular weights of the a and cy’ subunits predicted from the human cDNA clones are 45 160 and 41 400, which agree well with the molecular weight values of 45000 and 40 000 as determined by SDS-polyacrylamide gel electrophoresis of bovine testis or lung casein kinase I1 (Litchfield et al., 1990). This is an indication that there are no major co- or posttranslational modifications of the a and a’ subunits, i.e., proteolytic processing. There is some evidence for minor co- or posttranslational modifications. Isoelectric variants of both the a and the a’ subunits from rat liver nuclei and bovine testis casein kinase I1 have been observed by two-dimensional gel electrophoresis (Qi et al., 1986; Litchfield et al., 1990). Some of the isoelectric variation within each subunit appears to be due to phosphorylation since treatment with alkaline phosphatase changed the isoelectric pattern (Qi et al., 1986). The bovine lung a subunit appears to be blocked on a seryl residue (see Results), as is the /3 subunit of bovine lung casein kinase I1 (Takio et al., 1987). A blocked seryl residue is consistent with the data reviewed by Arfin and Bradshaw (1988) which suggest that eukaryotic proteins translated with a penultimate serine residue will lose the initiator methionine and that the serine residue will be blocked. Interestingly, proteins with proline in the penultimate position (like the a’ subunit) are usually not blocked (Arfin & Bradshaw, 1988). The existence of distinct, separately encoded a and a’ catalytic subunits raises the possibility of specialized properties and cellular functions. Those different properties may be conferred by the carboxy terminus of the two polypeptides, since this is the region of greatest sequence heterogeneity. Specialization could involve differential regulation of activity, substrate selectivity, subcellular localization, association with the (3 subunit, filament formation, or other properties of casein kinase 11. It is not clear whether the a and a’ subunits can exist independently of the @ subunit in vivo. There have been reports that casein kinase I1 purified from some sources is composed only of an a subunit (Baydoun et al., 1980; Renart et al., 1984) and that in vitro the a subunit has some catalytic activity in the absence of /3 subunit (Cochet & Chambaz, 1983b; Tuazon & Traugh, 1990). However, given the current understanding of casein kinase 11, any different properties or functions conferred by the a and cy’ subunits would be manifest in the holoenzyme. There are three possible isoforms: a2(32, The ratio of these isoforms would be deand pendent upon the expression of the individual subunits, and on the relative abilities of each of the subunits to form complexes with the others. From the Northern blot analysis of

Sequence of Casein Kinase 11 a and a’ Subunits RNA from A43 1 or HSB-2 cells (Figure 4), the mRNAs for both the a subunit and the a’ subunit are present. Some preparations of casein kinase 11 purified from homogeneous cultured cell populations contain both a and a’ subunits (Pyerin et al., 1987). These studies indicate that there are examples of a and a’ being expressed simultaneously in the same cell. There are also some reports of only one size of a subunit in casein kinase 11 (Boivan & Galand, 1979; Hathaway & Traugh, 1982; Cochet & Chambaz, 1983b; Glover et al., 1983; Meggio & Pinna, 1984) which suggests that there are examples of selective expression (or possibly only one gene in some species) of these subunits. Cochet et al. (1981) reported evidence suggesting distinct isoforms of the holoenzyme. The isoforms were distinct on the basis of elution from phosphocellulose resin, but not on the basis of subunit composition, substrate specificity, catalytic activity, or other parameters measured. However, the a subunits from both isoforms appeared much smaller than a subunits from other purified preparations of casein kinase I1 from the same organism (the bovine), suggesting proteolysis during purification. In our hands, analyses of casein kinase I1 holoenzyme rapidly purified from bovine testis have failed to resolve isoforms by size- or ion-exchange ~hromatography.~ Both the a subunit and the a’ subunit cDNA clones recognized multiple RNAs in samples of poly(A)-rich RNA (i.e., presumably mRNA). There are also multiple mRNAs for the subunit of casein kinase I1 (Heller-Harrison et al., 1989). These p subunit multiple mRNAs appear to differ in the 3’ noncoding region. A similar situation may exist for the a and a’ subunits of casein kinase 11. Messenger RNAs that have different noncoding regions could theoretically have different stabilities, different rates of nucleocytoplasmic transport, or different rates of translation, thereby affecting expression of the given subunit. It is also possible that the mRNAs may have different coding regions, for example, coding regions that encode signal or localization sequences that may or may not be proteolytically removed after being translated and processed through the appropriate transport pathway. Multiple mRNAs could be products of differential processing of a single RNA transcript, which can be generated by alternative splicing and/or the presence of more than one polyadenylation site [see Adams et al. ( 1 986)]. The multiple mRNAs for the p subunit of casein kinase 11 in HepG2 cells appear to be the result of more than one polyadenylation site (Heller-Harrison et al., 1989). Alternatively, some of the mRNAs could be the products of different, but highly related, genes. This remains a possibility because the analysis of genomic DNA cannot distinguish between the presence of a single gene with introns or the presence of an additional, highly related gene(s) for each subunit. Further analysis is required to determine the differences between the multiple mRNAs of the a and a’ subunits. To our knowledge, the only other organism (besides bovine and human) in which a and a’ subunits have been isolated and characterized is in the yeast Saccharomyces cereuisiae (Padmanabha & Glover, 1987; Chen-Wu et al., 1988; Padmanabha et al., 1990). The a and a’ subunits of S. cereuisiae (products of the CKAl and the CKA2 genes, respectively) are 60% identical, with most of the difference in primary structure between the subunits due to a 38-residue insertion in the a subunit. The primary structure of the a and a’ subunits in S . cereuisiae and in the human (or bovine) suggests that the a and a’ genes in each organism arose by independent gene D. W . Litchfield and E. G. Krebs, unpublished results.

Biochemistry, Vol. 29, No. 36, 1990

8445

duplication events (Padmanabha et al., 1990). This is reflected in the degree of identity of the S. cerevisiae subunits to the human subunits: CKAl vs human a, 69%; CKAl vs human a’, 71%; CKA2 vs human a, 64%; CKA2 vs a’, 64% [calculated after alignments of domains I through XI as compiled by Hanks et al. (1988)]. This sequence alignment and identity comparison do not reveal an obvious relationship of the two a subunits and the two a’ subunits. Instead, the two yeast subunits resemble each human subunit to a similar extent. The functions of the yeast a and a’ subunits have been addressed by genetic analysis. Disruption of the genes (by gene replacement) that encode the a (CKAl gene) and a’ (CKA2 gene) subunits is lethal for the resulting S . cereuisiae strain (Padmanabha et al., 1990). However, disruption of the CKAl gene alone or the CKA2 gene alone is not lethal; the strains with disrupted CKAl or CKA2 are normal with respect to growth on rich medium, resistance to heat shock, resistance to nitrogen starvation, and ability to mate (Chen-Wu et al., 1988; Padmanabha et al., 1990). The a and a’ subunits could have subtly different functions that were not detected in the limited analysis of the strains, but apparently, the a’ subunit can adequately compensate for the absence of the a subunit, and vice versa (Chen-Wu et al., 1988; Padmanabha et al., 1990). Therefore, in S . cereuisiae, the properties of the a and a’ subunits are sufficiently similar that either subunit can accomplish the essential cellular functions. Further experiments are required before similar conclusions can be made about the properties and cellular functions of the a and a’ subunits in mammalian cells. ACKNOWLEDGMENTS We thank Mr. Floyd Kennedy and Ms. Jenny Mulligan for assistance with the purification of bovine lung casein kinase 11, Ms. Maria Harrylock for assistance with sequencing the bovine lung a subunit peptides, and Dr. Patrick Chou, Dr. Yim-Foon Lee, and Ms. Rebekeh Shen (Chemical Synthesis Facility, Howard Hughes Medical Institute, University of Washington) for synthesis of oligonucleotides. We also thank Dr. Roger Perlmutter for supplying the human T-cell cDNA library and Dr. Claiborne V. C. Glover for the Drosophila cDNA clones and for copies of manuscripts prior to publication. Registry No. DNA (human clone hT4.1 casein kinase I1 a-subunit messenger RNA complementary), 128495-02-3; casein kinase I1 (human clone hT4.1 a-subunit precursor reduced), 128495-03-4; DNA (human clone hT9.1 casein kinase I1 a’-subunit messenger RNA complementary), 128495-04-5; casein kinase I1 (human clone hT9.1 a’-subunit precursor reduced), 128495-05-6; casein kinase 11, 52660- 18-1 ; casein kinase (human clone hT4.1 isoenzyme I1 deblocked a-subunit protein moiety reduced), 128495-06-7; casein kinase (human clone hT9.1 isoenzyme I1 a’-subunit protein moiety reduced), 128495-07-8.

REFERENCES Ackerman, P., & Osheroff, N. (1989) J . Biof. Chem. 264, 11958-1 1965. Ackerman, P., Glover, C. V. C., & Osheroff, N. (1985) Proc. Natl. Acad. Sci. U.S.A.82, 3164-3168. Ackerman, P., Glover, C. V. C., & Osheroff, N. (1988) J . Biol. Chem. 263, 12653-12660. Adams, R. L. P., Knowler, J. T., & Leader, D. P. (1986) The Biochemistry of the Nucleic Acids, pp 287-340, Chapman and Hall Ltd., London. Arfin, S . M., & Bradshaw, R. A. (1988) Biochemistry 27, 7979-7984. Baydoun, H., Hoppe, J., Jacob, G., & Wagner, K. G. (1980) FEBS Lett. 122, 231-233.

8446 Biochemistry, Vol. 29, No. 36, 1990 Boivan, P., & Garland, C. (1979) Biochem. Biophys. Res. Commun. 89, 7-1 6. Caizergues-Ferrer, M., Belenguer, P., Lapeyre, B., Amalric, F., Wallace, M. O., & Olson, M. 0. J. (1987) Biochemistry 26, 7876-7883. Carroll, D., & Marshak, D. R. (1989) J . Biol. Chem. 264, 7345-7348. Chen-Wu, J . L.-P., Padmanabha, R., & Glover, C. V. C. (1988) Mol. Cell. Biol. 8, 4981-4990. Cochet, C., & Chambaz, E. M. (1983a) Mol. Cell. Endocrinol. 30, 247-266. Cochet, C., & Chambaz, E. M. (1983b) J . Biol. Chem. 258, 1403-1 406. Cochet, C., Job, D., Pirollet, F., & Chambaz, E. M. (1981) Biochim. Biophys. Acta 658, 191-201. Dahmus, G . K., Glover, C. V. C., Brutlag, D., & Dahmus, M. E. (1984) J . Biol. Chem. 259, 9001-9006. Dang, C. V., & Lee, W. M. F. (1989) J . Biol. Chem. 264, 18019-18023. DePaoli-Roach, A. A., Ahmad, Z., & Roach, P. J. (1981) J. Biol. Chem. 256, 8955-8962. DePaoli-Roach, A. A., Ahmad, Z., Camici, M., Lawrence, J. C., Jr., & Roach, P. J. (1983) J . Biol. Chem. 258, 10702-1 0709. Edelman, A. M., Blumenthal, B. K., & Krebs, E. G. (1987) Annu. Rev. Biochem. 56, 567-613. Feige, J.-J., Cochet, C., Pirollet, F., & Chambaz, E. M. (1983) Biochemistry 22, 1452-1 459. Feinberg, A. P., & Vogelstein, B. (1983) Anal. Biochem. 132, 6-1 3. Feinberg, A. P., & Vogelstein, B. (1984) Anal. Biochem. 137, 266. Firzlaff, J. M., Galloway, D. A,, Eisenman, R. N., & Luscher, B. (1989) New Biol. 1, 44-53. Fourney, R. M., Miyakoshi, J., Day, R. S., & Paterson, M. C. (1988) Focus, Vol. 10, No. 1, pp 5-7, Bethesda Research Laboratories/Life Technologies, Inc., Gaithersburg, MD. Gatica, M., Allende, C. C., & Allende, J. E. (1989) FEBS Lett. 255, 414-418. Glover, C. V. C. (1986) J . Biol. Chem. 261, 14349-14354. Glover, C. V. C., Shelton, E. R., & Brutlag, D. L. (1 983) J . Biol. Chem. 258, 3258-3265. Gonzatti, M. I., & Traugh, J. A. (1988) Biochem. Biophys. Res. Commun. 157, 134-139. Goueli, S . A., Davis, A. T., & Ahmed, K. (1986) In?. J. Biochem. 18, 861-873. Grasser, F. A., Scheidtmann, K. G., Tuazon, P. T., Traugh, J. A,, & Walter, G. (1988) Virology 165, 13-22. Hanks, S. K., Quinn, A. M., & Hunter, T. (1988) Science 241, 42-52. Hara, T., Takahashi, K., & Endo, H. (1981) FEBS Lett. 128, 33-36. Hathaway, G . M . , & Traugh, J. (1982) Curr. Top. Cell. Regul. 21, 101-127. Hathaway, G . M., & Traugh, J. A. (1984a) J . Biol. Chem. 259, 701 1-7015. Hathaway, G . M., & Traugh, J. A. (1984b) J . Biol. Chem. 259, 2850-2855. Hathaway, G. M., Zoller, M. J., & Traugh, J. A. (1981) J . Biol. Chem. 256, 11442-1 1446. Hathaway, G. M., Tuazon, P. T., & Traugh, J. A. (1983) Methods Enzymol. 99, 317-331.

Lozeman et al. Haystead, T. A. J., Campbell, D. G., & Hardie, D. G. (1988) Eur. J . Biochem. 175, 347-354. Heller-Harrison, R., Meisner, H., & Czech, M. P. (1989) Biochemistry 28, 9053-9058. Herrmann, B. G., & Frischauf, A.-M. (1987) Methods Enzymol. 152, 180-183. Hindley, J., & Phear, G. A. (1984) Gene 31, 129-134. Inoue, A,, Tei, Y., Qi, S.-L., Higashi, Y., Yukioka, M., & Morisawa, S. ( 1984) Biochem. Biophys. Res. Commun. 123, 398-403. Jakobi, R., Voss, H., & Pyerin, W. (1989) Eur. J . Biochem. 183, 227-233. Kalderon, D., Roberts, B. L., Richardson, W. D., & Smith, A. E. (1984) Cell 39, 499-509. Klarlund, J. K., & Czech, M. P. (1988) J . Biol. Chem. 263, 15872-15875. Kozak, M. (1984) Nucleic Acids Res. 12, 857-872. Krebs, E. G., Eisenman, R. N., Kuenzel, E. A,, Litchfield, D. W., Lozeman, F. J., Luscher, B., & Sommercorn, J. (1 988) Cold Spring Harbor Symp. Quant. Biol. 53, 77-84. Kuenzel, E. A., Mulligan, J. A., Sommercorn, J., & Krebs, E. G. (1987) J . Biol. Chem. 262, 9136-9140. Lee, M. G., & Nurse, P. (1987) Nature 327, 31-35. Lees-Miller, S. P., & Anderson, C. W. (1 989) J . Biol. Chem. 264, 2431-2437. Litchfield, D. W., Lozeman, F. J., Piening, C., Sommercorn, J., Takio, K., Walsh, K. A., & Krebs, E. G. (1990) J . Biol. Chem. 265, 7638-7644. Littman, D. R., Newton, M., Crommie, D., Ang, S.-L., Seidman, J. G., Gettner, S. N., & Weiss, A. (1987) Nature 326, 85-88. Lorincz, A. T., & Reed, S. I. (1984) Nature 307, 183-185. Luscher, B., Kuenzel, E. A., Krebs, E. G., & Eisenman, R. N. (1989) E M B O J . 8, 1111-1119. Lyons, R. H., Ferguson, B. Q., & Rosenberg, M. (1 987) Mol. Cell. Biol. 7 , 2451-2456. Maenpaa, P. H. (1977) Biochim. Biophys. Acta 498,294-305. Mamrack, M. D. (1989) Mol. Cell. Biochem. 85, 147-157. Maniatis, T., Fritsch, E. F., & Sambrook, J. (1982) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Maniatis, T., Fritsch, E. F., & Sambrook, J. (1989) Molecular Cloning: A Laboratory Manual, 2nd ed., Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Marin, O., Meggio, F., Marchiori, F., Borin, G., & Pinna, L. A. (1986) Eur. J . Biochem. 160, 239-244. Meggio, F., & Pinna, L. A. (1984) Eur. J . Biochem. 145, 593-599. Meggio, F., Donella-Deana, A., Brunati, A. M., & Pinna, L. A. (1982) FEBS Lett. 141, 257-262. Meggio, F., Marchion, F., Borin, G., Chessa, G., & Pinna, L. A. (1984) J . Biol. Chem. 259, 14576-14579. Meisner, H., Heller-Harrison, R., Buxton, J., & Czech, M. P. (1989) Biochemistry 28, 4072-4076. Murao, S., Collart, F. R., & Huberman, E. (1989) J . Biol. Chem. 264, 8356-8360. Nakajo, S., Hagiwara, T., Nakaya, K., & Nakamura, Y . (1986) Biochem. Int. 13, 701-707. Okayama, H., Kawaichi, M., Brownstein, M., Lee, F., Yokota, T., & Arai, K. (1987) Methods Enzymol. 154, 3-29. Padmanabha, R., & Glover, C. V. C. (1987) J . Biol. Chem. 262, 1829-1835. Padmanabha, R., Chen-Wu, J. L.-P., Hanna, D. E., & Glover, C. V . C. (1990) Mol. Cell. Biol. 10, 4089-4099.

Biochemistry 1990, 29, 8447-845 1 Picton, C., Woodget, J., Hemmings, B., & Cohen, P. C. (1982) FEBS Lett. 150, 191-196. Proudfoot, J. M., & Brownlee, G. G. (1976) Nature 265, 21 1-214. Pyerin, W., Burow, E., Michaely, K., Kubler, D., & Kinzel, V. (1987) Biol. Chem. Hoppe-Seyler 368, 215-227. Qi, S.-L., Yukioka, M., Morisawa, S . , & Inoue, A. (1986) FEBS Lett. 203, 104-108. Renart, M. F., Sastre, L., & Sebastian, J. (1984) Eur. J . Biochem. 140, 47-54. Richardson, W. D., Roberts, B. L., & Smith, A. E. (1986) Cell 44, 77-85. Rose, K. M., Bell, L. E., Siefken, D. A,, & Jacob, S . T. (1981) J . Biol. Chem. 256, 7468-7477. Sanger, F., Nicklen, S . , & Coulson, A. R. (1 977) Proc. Natl. Acad. Sci. U.S.A. 74, 5463-5467. Saxena, A., Padmanabha, R., & Glover, C. V. C. (1987) Mol. Cell. Biol. 7, 3409-341 7. Schneider, H. R., & Issinger, 0. G. (1988) Biochem. Biophys. Res. Commun. 156, 1390-1 397.

8447

Sommercorn, J., & Krebs, E. G. (1988) in Advances in Post- Translational Modifications of Proteins and Aging (Zappia, V., Galletti, P., Porta, R., & Wold, F., Eds.) pp 403-41 5, Plenum Publishing Corp., New York. Sommercorn, J., Mulligan, J. A,, Lozeman, F. J., & Krebs, E. G. (1987) Proc. Natl. Acad. Sci. U.S.A.84, 8834-8838. Takio, K., Kuenzel, E. A., Walsh, K. A., & Krebs, E. G. (1987) Proc. Natl. Acad. Sci. U.S.A.84, 4851-4855. Thornburg, W., & Lindell, T. J. (1977) J . Biol. Chem. 252, 6660-6665. Thornburg, W., Gamo, S., O'Malley, A. F., & Lindell, T. J. (1979) Biochim. Biophys. Acta 571, 35-44. Tuazon, P. T., & Traugh, J. A. (1990) Adu. Second Messenger Phosphoprotein Res. (in press). Witters, L. A,, Tipper, J. P., & Bacon, G. W. (1983) J . Biol. Chem. 258, 5643-5648. Witters, L. A., Watts, T. D., Daniels, D. L., & Evans, J. L. (1988) Proc. Natl. Acad. Sci. U.S.A. 85, 5473-5477. Zandomeni, R., Zandomeni, M. C., & Weinmann, R. (1988) FEBS Lett. 235, 247-251.

Modification of Bases in DNA by Copper Ion-1, 10-Phenanthroline Complexes+ Miral Dizdaroglu,*i* Okezie I. Aruoma,sJ and Barry Halliwells Center for Chemical Technology, National Institute of Standards and Technology, Gaithersburg, Maryland 20899, and Biochemistry Department, University of London King's College, Strand Campus, London WC2R 2LS. United Kingdom Received March 14, 1990; Revised Manuscript Received May 30, I990

ABSTRACT: Damage to the bases in D N A by the cupric ion-1,lO-phenanthroline complex was investigated. Ten base products in D N A were identified and quantitated by the use of gas chromatography/mass spectrometry with selected-ion monitoring. D N A damage by the cupric ion-1,lO-phenanthroline complex required the presence of a reducing agent such as ascorbic acid or mercaptoethanol. Products identified were typical hydroxyl radical induced products from the pyrimidines and purines in DNA, well-known from previous studies using various hydroxyl radical producing systems such as ionizing radiation, hypoxanthine/xanthine oxidase, or hydrogen peroxide in the presence of transition metal ions. Product formation was not significantly inhibited by typical scavengers of hydroxyl radical such as mannitol and sodium formate, but there was partial inhibition by dimethyl sulfoxide. Catalase substantially decreased formation of base products, and added hydrogen peroxide stimulated it, indicating the hydrogen peroxide dependency of D N A base damage. Superoxide dismutase afforded only a partial reduction in product yields in systems containing ascorbic acid. O n the basis of the types of base products formed, the hydrogen peroxide dependency of product formation, and a previous report suggesting that D N A damage is due to a diffusible species [Williams, L. D., Thivierge, J., & Goldberg, I. H. (1988) Nucleic Acids Res. 16, 11607-116151, we propose that D N A base damage is caused by hydroxyl radical.

In

the presence of H202,a 2:l complex of the chelating agent 1,lO-phenanthroline(OP)' with Cu+ ions [(OP),Cu+] produces strand cleavage of DNA after binding to the minor groove (Sigman et al., 1979; Sigman, 1986). The nuclease activity of the (OP)2Cu+ complex has been widely used for structural studies upon DNA (Sigman, 1986; Marshall et al., 1981; Pope & Sigman, 1984; Veal & Rill, 1988). In addition, measuremencof DNA damage in the presence of an excess of DNA and of OP has been employed as a sensitive assay to measure We are grateful to the Association for International Cancer Research, Uxbridge, U.K., for partial financial support. * T o whom correspondence should be addressed. *National Institute of Standards and Technology. 5 University of London King's College. Part of this work was done when O.I.A. was a guest scientist at the National Institute of Standards and Technology.

the availability of copper ions in human body fluids (Gutteridge, 1984; Gutteridge et al., 1985; Evans et al., 1989). During experimentation, Cu2+ions are often mixed with the O P so that a reducing agent has to be added to reduce Cu2+ ions to Cu+ ions and form the DNA-cleaving species. The reducing agents that have been employed in vitro to facilitate I Abbreviations: OP, 1.10-phenanthroline; 02-, superoxide radical; 'OH, hydroxyl radical; GC/MS-SIM, gas chromatography/mass spectrometry with selected-ion monitoring; 5-OH-5-MeHyd, 5-hydroxy-5methylhydantoin; 5-OH-Hyd, 5-hydroxyhydantoin; 5-OHMeUra, 5(hydroxymethy1)uracil; 5,6-diOH-Cyt, 5,6-dihydroxycytosine; FapyAde, 4,6-diamino-5-formamidopyrimidine; FapyGua, 2,6-diamino-4-hydroxy5-formamidopyrimidine; 8-OH-Ade, 8-hydroxyadenine; 8-OH-Gua, 8hydroxyguanine; ME, mercaptoethanol; SOD,copper-zinc superoxide dismutase; asc, ascorbic acid; TBA, thiobarbituric acid; Me2S0, dimethyl sulfoxide.

0006-2960/90/0429-8447$02.50/00 1990 American Chemical Society